Research On Fault Location In Distribution Network With A Single-phase To Ground Fault Based On Hilbert-huang Transform And Genetic Algorithm

The precise fault location in the distribution power system network is a difficult problem, which has not been solved well.Studying on the problem has an important role in maintaining and restoring the system when a fault occurs. With the development of automation technology in distribution network and increasing requirements of power supply quality of users, the existing methods of fault location are not very effective in distribution network, it is necessary to further study the new method in distribution network.As for the existing methods of fault location are not very effective in distribution network, a new method based on non-power frequency transient component to select the fault line and a new method based on Hilbert-huang transform and genetic fault characteristic matching technology of fault location in distribution is proposed.Hilbert-huang transform is used to decompose fault zero-sequence current singals, to extract non-power frequency components of each line. The non-power frequency current of the fault line, extracted with HHT, have a contracy polarity as the currents of the other non-fault lines at the one-tenth of the wave period time after the fault occurs. These characteristic are used as evidence for fault selection in distribution network with the non-power frequency transient component. The method is precise and reliable.Based on the precise fault line selection, Hilbert-huang transform is used to extract the power distribution characteristics of zero sequence fault current in different time with different frequency. The genetic algorithm is used to automatically search the most similar simulation waveforms to the actual fault waveforms under different fault distance, grounded resistance and fault time.The fault position in the simulation condition of the most similar power distribution of simulation waveforms is the output of the fault distance. The new method using non-power frequency component to select the fault line, Hilbert-huang transform to extract the characteristic and genetic algorithm to automatically search the similar simulation waveforms is formed, and it has an important role in improving the precision of fault location.The new method of fault location in distribution network with a single-phase-to-earth fault based on Hilbert-huang transform and genetic algorithm can be used to locate the fault point precisely, and can be used in air-line, cable and cable-line un-effectual grounded system.